00:00 - 15:0015:00 - 00:00
3:14 PM
For the record, and for context, Smoke Detector is open source, and thus it does not have an "evil creator mastermind" in the sense you seem to be alluding to. The users who were responsible for the original architecture are by and large no longer very active in the Charcoal community. For the terminally curious, the Github commit history at github.com/Charcoal-SE/SmokeDetector has the full details. — tripleee 53 secs ago
3:25 PM
@tripleee that was meant to be satire. If suddenly the company bans usage of AI generated content detector as a support tool to facilitate moderation because they are unreliable because they are just heuristic/probabilistic tools, then they have to ban Smokey too since on principle it works in the same exact way. Basically I am arguing that if they were fine with users helping themself with Smokey for years, it feels quite weird to ban usage of similar tools to help find low quality ai generated content. — SPArcheon 17 secs ago
I've tried to focus your title on your specific problem. Feel free to adjust it if you feel it doesn't do your intent justice. — Spevacus 54 secs ago
Not my downvotes, but this doesn't tell us anything new; it is merely an example of how ChatGPT is sometimes able to provide by and large correct answers to questions. These techniques and heuristics are well-known to anyone who has started to inform themselves about this issue, and not really useful in the sense that they would provide a robust, scalable solution to the problem. If it wasn't meant as a joke, it doesn't particularly contribute anything useful to the discussion. If it was, rest assured we have heard it many times already. — tripleee 10 secs ago
« it's written by a high-reputation user». This is not a reason to upvote. It is unfair, incorrect, user-based voting, and harms the site. — Andreas detests censorship 35 secs ago
@Andreasdetestscensorship i mean. that doesn't change the fact that people do it. This answer isn't suggesting it's a good practice. — Kevin B 21 secs ago
I've found it very useful in generating example sentences using Spanish words for my anki deck. — SurpriseDog 36 secs ago
Ask a question on the answer itself and how it relates to your issue and state that you don't understand the answer nor how to manipulate it to your issue, link back to it using the
share button on the answer. That is how we do it on TeX.SE — JamesT 21 secs agoWait.... WHAT??????? " the day that the announcement was posted for moderator “head’s up” was a holiday in the United States". Wasn't that exactly one of the "coincidences" that happened to Monica too??? — SPArcheon 44 secs ago
@SPArcheon For Monica it was specifically a religious holiday that, based on her religious views, specifically forbid her doing the sort of work that participating here entails. So, that was really far worse, though also from what I know was done out of ignorance rather than malice. — Bryan Krause 5 secs ago
@ThomasOwens You might consider altering the double-negative in "I believe that not taking strong, decisive action against algorithmically-generated content is the wrong thing to do" to instead read that you believe taking strong action is the right thing to do, though it does change the meaning a bit and if you think it's more important to highlight the wrongness of this policy rather than the rightness of the policy you prefer then it's certainly worth keeping the way it is. — Bryan Krause 18 secs ago
@BryanKrause once is ignorance, twice is suspicious, trice is malice. Hopefully we don't get to act three. — SPArcheon 29 secs ago
It does more than just mark it as resolved, yes, but one of its impacts is just that. Your reply to my second comment seems to misinterpret it. — CDR 27 secs ago
4:05 PM
@tripleee "well-known to anyone who has started to inform themselves about this issue" That's also what I see as the greatest strength of ChatGPT, bringing people up to date with the basics of a topic. It cannot replace an expert solving a difficult problem, but maybe it can be the initial first responder to questions of beginners. Helping them to find out what they really want. There are legitimate usecases for AI, looking over contributions and giving improvement tips, guiding people towards canonical answers, explaining basic concepts. All this would be useful. — Trilarion 45 secs ago
I wonder if the mods themselves have somehow gauged the effectiveness of the moderation of the AI content so far? Or are we all flying blind to some degree? — Trilarion 27 secs ago
The existing content remains valuable, so the site will probably not turn into Yahoo!.com any time soon. Expect to see the company get repeatedly sold off until it finally lands with someone who only cares about the traffic and replaces the contents with TV Shop ads. The path there will be erratic. There is no way to predict the trajectory or the velocity, only that the spasms of desperate changes will increase in frequency. — tripleee 11 secs ago
A bias in flags for certain countries/regions is not specificaly bound to AI-related flags. Removing those flag won't solve the bias for other flags. — The_spider 41 secs ago
What would make more sense would be to prohibit the OP from voting to reopen their own post. The vast majority of users who click "This edit resolves the close reason" are lying, knowingly or unknowingly. — tripleee 26 secs ago
@tripleee - it hurt my OCD so much I fixed it, long before I got to your comment. I anglicised too; forums is fine (I also noted in the edit summary that "Stack Exchange is not a forum"™ ;) — Tetsujin 5 secs ago
The new design unfortunately does not let me downvote this post 1.28 times. Nothing else is circular except for the history icon. These buttons look so out-of-place in the page. — Khuldraeseth na'Barya 26 secs ago
@Andreasdetestscensorship I think that he actually means: "Due to these types of voting, voting alone can't be relied upon" - that's what I understand from it — Example person 45 secs ago
Been on the site almost daily for +10 years. Of all the "improvements" the SO geniuses have done over time, this is defently the worst. I will never like black circles with deep black arrows, and blue / orange instead of green / red. — davidkonrad 54 secs ago
@Andreasdetestscensorship I completely agree, I hope the edit makes things clearer? — Glorfindel ♦ 16 secs ago
I don't have enough words to describe how unbelievably out of touch that particular blog post is. — Oleg Valter is with Ukraine 34 secs ago
String composition for translations always runs the risk of the resulting paragraph not making a ton of sense, but I think in this case it's reasonable to split those out. — Adam Lear ♦ 43 secs ago
Hi Marco Gadda, welcome to the Stack Exchange Network Meta site! I'm not sure which search brought you here but the problem you describe will not be answered on this specific site. To get an expert's answer for the topic of your question you'll have to find and then re-post on the proper site. Check How do I ask a good question and What is on topic on the target site to make sure your post is in good shape. Your question is definitely off-topic on Meta and is better deleted here. — Spevacus 1 min ago
@Trilarion Of course we have. We care about our sites and are constantly self-evaluating our actions and the actions of our fellow mods. We don't earn any sort of points for the most accounts suspended or most posts deleted or anything like that. All indications to moderators are that the vast majority of actions that mods take against AI-generated content have been correct to community standards, and we haven't been provided with a single example that suggests otherwise. — Bryan Krause 19 secs ago
SO for Teams is for companies, like as an internal knowledge base for questions specific to that company or an internal tool — cocomac 25 secs ago
I fully understand the demand of having public data on the risks of misdetecting AI generated content. A risk that seems motivated well since AI is trained to mimick human generated content. Detecting AI generated content seems to be a non-trivial task. However, I wonder what mods will do if the data is not forthcoming or not convincing enough? In the end it always comes down to deciding if under the current policies it's worth moderating or not. And the company should have some freedom in setting these policies, after all it's their site. — Trilarion 11 secs ago
"3- Some advice on how tò implement them on my website made with WordPress." How is this relevant to Stack Exchange? — Mast 36 secs ago
@BryanKrause I see. So basically it seems that the perceived risk of misdetection differs greatly: the company thinks it's really big, the mods think it's really small. Both groups have somehow checked their data, but none is presenting any of that to the public. For us users it comes down to whom we trust more. I think it's clear how this competition of trust ends just wanted to make it as clear as possible. For the mods the question is what to do if the company cannot be convinced that AI generated content can be detected without much risk? Am I getting it approximately right? — Trilarion 53 secs ago
@Glorfindel Yes, definitely. It's fine to point out these reasons, because they are real, but they should be avoided. — Andreas detests censorship 35 secs ago
@Trilarion I think the biggest disconnect is in the assessment of different risks. The community consensus, shared by moderators, is that AI-generated content poses substantial risks to the quality of content here, especially due to the ease of producing such content relative to the ease of curating it. The company has alluded to internal evidence that overmoderation of this content is harming the site by discouraging participation. I have not seen any evidence that this is the case that does not have far simpler explanations. — Bryan Krause 8 secs ago
It's definitely kinder than the one I submitted ("You’re a liar. You do not have our (the community’s) trust!"). I do admire your steadfast determination to continue explaining and presenting things nicely, but I'm surprised you haven't just given up. They are clearly not listening, and have no idea what they're doing. — Andreas detests censorship 18 secs ago
@CDR, unless I am getting you wrong, I think you are missing my whole point: I know one of its impacts its just that, I say it myself in the question. What I note is that this effect could be achieved with a feature that ONLY does that, namely signaling the question doesn't need more replies, without highlighting a specific answer---and hence avoiding the vices related to this feature---. — Albert 16 secs ago
I am seeing relative links; not sure whether this is a change since you've posted this or if something else is causing you and I to see things differently. — Bryan Krause 55 secs ago
This makes sense for existing content, but enacting punishments to users for them would be unusual, save for still recent situations. For instance, back in 2018, we were suggested not to use unfriendly/unkind flags for old borderline comments, as they were likely not to be representative of the user's latest conduct. — E_net4 28 secs ago
@Mast oh, I do have a couple of those too. And they are against any imaginable CoC too... — Oleg Valter is with Ukraine 57 secs ago
Thanks - that was a quick fix then. I'm trying to find a Wayback Machine copy of a page to convince it was different when I posted this :) — Glorfindel ♦ 19 secs ago
@ThomasMarkov either that, or that the CEO doesn't have any clue about what's happening right now. Or, which is more likely, doesn't give a flying <insert expletive> about it. — Oleg Valter is with Ukraine 32 secs ago
Please take the time to take the Tour and read the Help pages, especially What topics can I ask about here? — Tetsujin 32 secs ago
If the content was from another website, one would simply link to that site or even argue with fair use if the snippet was only small. Just for clarity: why would linking to ChatGPT not be enough (if ChatGPT wasn't banned)? Or is the issue simply that people cannot circumvent the ban without not giving attribution? — Trilarion 46 secs ago
Well the reason for not asking for feedback this time is obvious, and I find it fair. They are well aware how it would be received, and afraid of it, so prefer to just throw the bomb and run to hide behind a shelter. — Shadow The Spring Wizard 36 secs ago
"What are your thoughts..." -- unpublishable (if I don't want to get banned). — Dan Mašek 28 secs ago
@ShadowTheSpringWizard I don't find it fair at all. If you're going to publish something that offends a bunch of people, they're going to have some viewpoints on that, hence why I posted a discussion post even though SE didn't. Especially after the recent rule change, I wasn't very inclined to allow them to try and sweep the blog post under the rug — cocomac 1 min ago
We could try to get generative AI to stop hallucinating facts. Or we could just get CEOs to hallucinate facts as well, to level the playing field. The easiest way to get an AI with human-level performance is obviously to decrease human-level performance, and Stack Exchange seems to get that. Good to know they're clearly on top of it. — Erik A 8 secs ago
So, the problem with making every so scattered (so many questions the last few days), is that's it hard to feature them. Which one should be featured? Which one should not? We're failing at making the community visible in this. — Andreas detests censorship 46 secs ago
I agree with many content points and I understand that you are upset but I also find the wording in this answer borderline abusive. The aim should still be a respectful discourse, no matter what we perceive the other side to be. Getting angrier and angrier won't be the solution. — Trilarion 35 secs ago
I see. Your title reads more like one for discussion, consider changing it to one more reflective of your post's nature. Perhaps something along the lines of 'Give users a way to mark their question as resolved without accepting an answer' would be more descriptive. (Just a suggestion.) — CDR 1 min ago
I think it's just the usual business speak. Almost empty phrases with low density information in between. One could probably summarize to: we fired people and we want to make money with AI. It's not more really. If only blog posts wouldn't be so long. — Trilarion 28 secs ago
Whenever I see 'CEO' in the question title nowadays I flinch. This guy seems to have no clue as to what we as a community want/no care as to what we as a community want. — CDR 49 secs ago
@snakecharmerb IMHO there are few flawed assumptions there: a) that a bunch of career lairs and manipulators see any value in truthful and factual content. b) that these people are actually interested in building something of lasting value (as opposed to sucking it dry and discarding the empty husk when they inevitably move on to ruin something else). — Dan Mašek 39 secs ago
@Trilarion Consider for a moment that I am correct in my assessments; you’re then asking me to show respect to an abuser, and allow myself to be abused without protesting. — Andreas detests censorship 42 secs ago
@BryanKrause Ok so let's hope that the disconnect can be mended. Because if not and that is not unlikely, something has to give. The irony is that I agree that ChatGPT is currently not healthy for SO but at the same time I'm using it almost daily for my work because it's useful to me. But then I know what I'm doing and can compensate for the shortcomings of ChatGPT. — Trilarion 42 secs ago
@Trilarion I think there are even some promising integrations of ChatGPT with SE/SO; Philippe posted recently a couple possible ways to go, which I think are all workable with some refinement. But intermingling actual content generated by AI with human-generated content is something I believe is strongly damaging, and makes it harder to use the sort of workflow I expect to develop, where people like you may use ChatGPT for some things but still need to come to SO when it breaks or they get stuck. I think ChatGPT is also not as useful for beginners who don't know how to spot its mistakes. — Bryan Krause 17 secs ago
"features [...] leveraging GenAI that have the potential to increase engagement within our public community". Well, it might, but not with human participants or anyone who has any idea what's really happening here. — Laurel 35 secs ago
@Trilarion There are multiple problems with GenAI content. One of them, the one addressed by this part of the CoC, is lack of attribution. The TOS of some of these programs already requires attribution when the generated content is used, but even without that, we require attribution whenever your writing is not your own. Whether acknowledging that ChatGPT generated content is sufficient or not remains somewhat unclear, especially because GenAI is currently incapable of attributing its own content. That's a problem if it effectively regurgitates training data in some contexts... — Bryan Krause 54 secs ago
...It's also a problem with reliability of information, which is more important on some sites on the network that have specific requirements for referencing answers, including Medical Sciences, one of the sites I moderate. That issue seems less problematic under the current guidance, because answers that fail to reference sources are not allowed regardless of who wrote them. The specific issue with the new policy, though, is that even if a post is apparently AI-generated and does not give proper attribution, we are forbidden from acting on it. — Bryan Krause 52 secs ago
Not just difficult on SO meta too, arqade meta has the same problem. Pretty much impossible to see if I've voted at a glance 🙄 — GammaGames just now
@tripleee Depends on your point of view, context and the situation. I had a number of full-scale edits (to actually resolve the off-topic issue) and my question remained closed after I: (a) checked the magic checkbox, (b) vote to reopen, (c) called of mod, (d) wrote a meta question. All for nothing, so I'd say that this is highly depending on the scenario. — trejder 47 secs ago
“I'm just confused. One moment they ban AI and the next moment they welcome it as a holy grail.” — Stack Overflow Inc (the company) never banned AI. We the users did on various sites. When it was banned on Stack Overflow (the site), that was by the volunteer moderators who are users like you and me. We requested SO Inc (the company) ban it network-wide. SO Inc declined to do so then and is welcoming it now. SO Inc's position has been consistent, but also consistently opposite the community. — doppelgreener 59 secs ago
"Then stop insulting us, lying to us, sabotaging us, and generally treating us like shit." -- spot on, although I'm afraid that request is about as likely to be fulfilled as if you asked them to stop breathing. Sadly, for some people, that's just the way of life. — Dan Mašek 30 secs ago
"Emphasis is mine, and I'm pleased to see you admit what your only goal is." you're misrepresenting what the blogpost actually wrote. I'm not a fan of the direction Prashath is leading things in, but I personally won't misquote him to mudsling. — starball 11 secs ago
@starball In fact, you may actually be correct about that! I’ll have a closer look once I get home! Thanks! — Andreas detests censorship 17 secs ago
Does this answer your question? It’s hard to tell how I’ve voted on meta sites. Please make selected and unselected vote buttons different colors — honk 13 secs ago
"So in essence, you want to use us as free labour to develop a product you'll make lots of money off from, one of which we get nothing in return from; only burnouts." - the network sites making profit for them is not new. And it's not fully true to say that everyone here gets "nothing" from this and only burnouts- at least- I am not confident that that is a fact. I'm not saying people aren't getting burned out, but not everyone is. — starball 50 secs ago
In my view, you tend to exaggerate quite heavily in your writings and use charged language. I'm not confident that that's the most effective way to get SE Inc. to listen. But to each their own. — starball 14 secs ago
@honk That one is about child meta sites. This seems to be about non-meta. — Bryan Krause 50 secs ago
@Catija The by-design version is also hard to see, as many people have pointed out here: meta.stackexchange.com/questions/389359/… — Bryan Krause 35 secs ago
Re: "Misc Suggestions on Links" – when reading the new Code of Conduct, I tried looking up "Acceptable Use policy" in the Help Center, and it came up with nothing relevant, so I'm very pro more linking to relevant pages. — andrybak 8 secs ago
@BryanKrause: True, but it's the same problem. I guess we have a general problem here ;) — honk 28 secs ago
I would tweak "Harmful political content policy", it sounds vague and contradictory. Like, it states that it may allow criticism as long as they do not otherwise violate this Code of Conduct, however it's also stated that it's forbidden to spread political misinformation or widely disproven allegations not supported by reasonable evidence to be promoted on the platform, but how do you even define reasonable evidence (I mean, it sounds for me, as a non-native speaker, as if you can criticize one people but not another, though in theory this limitation doesn't really make sense to me)? — nicael 38 secs ago
(disclaimer: I may just misinterprete the following wording) - all these questions aside, why should any politics "be promoted on the platform"? There's not even a place for any politics to be promoted at all expect for specific SE sites which are directly connected to politics (such as Law or Politics) — nicael 58 secs ago
@honk Indeed, the voting here: meta.stackexchange.com/questions/389359/… would indicate others also see the general problem. — Bryan Krause 15 secs ago
@nicael Don't forget chat rooms. That's a driving factor behind some of those things that may be allowed in some places. There is also content that can be put into user profiles where things that may not be allowed on Q&A would be allowed. But I do generally agree. — Thomas Owens 48 secs ago
The users using ChatGPT and other AI to get answers are going to steal all our reputation and activity and they will get everything. AI isn't the future. human minds are the future! (Collectively) — Big Joe 8 secs ago
I think the idea is that a lot of people see them soliciting feedback and think "oh those guys are great", whereas only few will actually go through the effort of checking whether they acted upon it (and be miffed when they find they didn't). And they really seem to just care about the numbers these days... — Dan Mašek 55 secs ago
@BigJoe - Quora now has a ChatGPT box front and centre directly above every single human-written answer. — Richard 46 secs ago
@Trilarion then we shouldn't do such a heavy handed approach. Guilty until proven innocent is bound to piss off many more people and with reason. — Braiam 6 secs ago
The goal is to increase the number of votes. Perhaps this increases the number of votes by forcing you to click it over and over to even tell if you voted? — Chris - Regenerate Response 12 secs ago
@Chris-RegenerateResponse I think I've (unintentionally) retracted more votes than I've cast in the last week... — Radvylf Programs 22 secs ago
"simultaneously implying that mods were somehow moderating this content without SE's awareness of what was being done and how it was being done" well, how are post being evaluated? If SE doesn't know how moderators do it, at least the community at large should be able to be privy of such process of evaluation. — Braiam just now
@RadvylfPrograms I've done it more than once. It's even harder to tell on some sites than others. — Chris - Regenerate Response 19 secs ago
"simultaneously implying that mods were somehow moderating this content without SE's awareness of what was being done and how it was being done" well, how are posts being evaluated? If SE doesn't know how moderators do it, at least the community at large should be able to be privy of such process of evaluation. — Braiam 1 min ago
Didn't realize that @Richard I can't believe all the question-and-answer sites are doing this. Quora, SE, Reddit killing off third-party apps. WTF is really going on. It looks like all these companies are looking at their bottom line and what the investors want. The Execs don't realize that we the people are what drive their traffic. without us there is nothing! — Big Joe 40 secs ago
One consideration might be that anyway not that many people will read the CoC. Surely not a lot of all those new members. The impact might be small either way and it comes down to how it's lived and enforced in practice. There are advantages to being short but I guess that companies rather tend towards some kind of "legislation". That's the result. One could probably check the length of each iteration of the CoC. — Trilarion 37 secs ago
Potentially it should work for all the scenarios, from grayscale version to the night mode (in the latter case if the site background is dark, the active voting button should have a white background) — nicael 42 secs ago
@JonCuster Of course, and the sarcass here is that ChatGPT probably dig all sites including SO/SE, etc.. to write up the text. You allow to be index, and you get killed of it.. and too late to block crawler, there is already a ton of honey trap mirrors of the whole site online. — yagmoth555 15 secs ago
8:12 PM
The CEO will have had nothing to do with that post. He probably has no idea what's happening, as you say, but he also didn't actually write that post. — terdon 31 secs ago
@Chris-RegenerateResponse Yes! The next step is to allow for multiple votes on the same content, and we're done! — Joachim 9 secs ago
Why should it be in correct English?? it does not sound nice, when communities from all over the world are participating the knowledge why it is for English speaking only?? I have see this many time that people with poor English ask the question and they get down vote, this not a place to correct and beautiful language you have to define your primary goal, question is important upper and lower case are not that important, for sure its good to write in a correct manner, positive way could be to correct if you see any mistake, we can't identify the cause of poor wr we should not make it a base. — sairfan 1 min ago
@Andreasdetestscensorship attacking and calling them names has also never worked, why would you continue doing that? If anything does have a chance of getting through, it would be a civil, constructive message like Bryan's. If we all go and give them a piece of our mind, we will quite rightly be ignored as trolls. Just because they are being deceitful and insulting doesn't mean we should be as well, after all, we're claiming the moral high ground here and behaving badly makes us lose it. — terdon 20 secs ago
Teams is the money-making part of the company, so that much makes sense. What doesn't make sense is breaking the public sites. One of the big selling points for Teams is that people who are familiar with and have a favorable view of the public SE/SO sites are going to be familiar with the Teams interface that looks and functions a lot like SE/SO. The reputation that they sell Teams on is the reputation of the public network. We're not the customers, we're part of a product they are selling, but it's potentially a collaborative relationship where we both gain. — Bryan Krause 25 secs ago
Posted a seemingly universal solution to a broader discussion: meta.stackexchange.com/a/389672 — nicael 33 secs ago
Does the "Dehumanization" section at stackoverflow.com/conduct/abusive-behavior now mean that the phrase "polishing a turd" is a violation? — TylerH 41 secs ago
Might the countries thing also potentially have something to do with the fact stack overflow reputation has a real monetary value in some places, it can be a precondition for work, providing more incentive to use things like ai. — user1937198 26 secs ago
Suppose you follow the middle ground, what happens if you remove step 1? What value does chatgpt actually give you? — user1937198 19 secs ago
@user1937198 Quite a bit, IMHO - ChatGPT can be extremely helpful in generating ideas on how to solve something, as long as you have the expertise to know when it's tossing out good suggestions and when it's just completely hallucinating. — NotTheDr01ds 10 secs ago
If it's just ideas, go back and find a source for the idea elsewhere afterwards, you need so for the understand phase. — user1937198 34 secs ago
@TylerH It does not, unless somebody's goal is to produce posts that are worthless (but we can safely presume this is nobody's intent, absent a statement of intent to the contrary). The only behaviour described by that phrase is "polishing". However, you should avoid calling any specific person's question or answer a "turd": please restrict use of that phrase to the general phenomenon of people copyediting deletion-worthy posts. — wizzwizz4 40 secs ago
The public version didn't address anything significant from the moderators' private feedback, either. Such a democratic process here from the SE community team. — Alexander Gruber 17 secs ago
@wizzwizz4 Yes, I meant in reference to a specific post, e.g. "don't bother editing that post you linked to; it's just polishing a turd". Though it seems... hollow?... to ban one use and not the other, frankly. — TylerH 48 secs ago
@TylerH That wouldn't be a violation of the "dehumanization" policy, though; it'd be a violation of "don't compare people's work to human excrement" policy (formerly "be nice"). — wizzwizz4 42 secs ago
@TylerH The "Engaging with users" section: «No matter where you engage on the network with your peers, we expect all users to treat one another with kindness and respect.» — wizzwizz4 27 secs ago
Thank you for bringing this to our attention. Our team is working on updating the colors of the button after voting on Meta sites so it will be much more visible. — Sasha ♦ 53 secs ago
There are many posts across the SE network that go without votes of either kind and, as mentioned in the post, voting has been declining year over year. This makes it harder for users to find value on the platform because they are less able to quickly judge the quality of the content. Encouraging more voting supports curation across the network, making it easier for users to locate high quality questions and answers. — Sasha ♦ 10 secs ago
Maybe they should include a list of examples under each rule to clarify things... — TylerH 30 secs ago
@wizzwizz4 The dehumanization policy on the page I linked to says it includes "comparing humans, groups, or their stated or perceived behaviors [...] such as [...] filth". I would consider a question I've asked to be part of my behavior because it is an action I have taken (which is what behavior means). — TylerH 58 secs ago
Do you have examples of "legitimate users" who were driven away by being wrongly suspended for ChatGPT content? The ChatGPT content I've seen hasn't exactly been stellar quality. For example, I got a ChatGPT answer on one of my questions that didn't answer the question in the least. — EJoshuaS - Stand with Ukraine 41 secs ago
@TylerH A person can produce filth without being filth: all they'd need to be is biological. (Extradiegetic reason: insulting someone's work is not what "dehumanization" means, and I'm fairly confident the CoC intends to use words normally.) That being said… this is one of those cases of "if you're asking, don't do it". Most people don't have the presence of mind to think about their actions on the meta-level like that: use this rare power for good, not evil! — wizzwizz4 33 secs ago
No one wants this, no one asked for this, no one finds this useful, no one can confirm your data. Everyone wants other things, everyone sees true problem (like area51 for example), everyone is asked for things that everyone wants but are declined with out (good) reason, and nearly everyone is upset about stack exchanges management. This is one of the worst series of things stack has done (The Monica cellio thing was worse) — Starship is go for launch 38 secs ago
@wizzwizz4 Sure, I'm not talking about calling a person filth. I'm talking about referring to their behavior as filth by saying their question is "a turd" indirectly via the suggestion that editing their question would be "polishing a turd". I don't generally say that but it is a somewhat time-honored saying on Stack Overflow (like help vampires, which was another term that was once popular but is now frowned upon/disallowed). — TylerH 48 secs ago
Stack Exchange is stupid, what can you do...Short of Monica resistance style mass leaving... — Starship is go for launch 54 secs ago
@TylerH As was "rep whore". I think of "help vampire" like "techbro": useful for describing an aggregate group of people, but should only exceptionally be tolerated when used to describe individuals. "Polishing a turd" is different, but it feels like the same logic applies. — wizzwizz4 just now
you have little to no control over how your content will be applied by people outside of SE if they follow the CC-BY-SA license. People can use what you write to make weapons of mass destruction. When you made an account, you implicitly agreed to the ToS, which also gives SE the right to use your content for commercial purposes. — starball 46 secs ago
I am aware of that. But what I am saying is I dont want to post more content to be fed to an AI. I know they can, but if I dont like what they do with it, I dont have to keep posting @starball — Starship is go for launch 22 secs ago
it's there website they can do ehateyrhry want. they are supposed to be transparent but obviously something else is happening here — Big Joe 49 secs ago
@Glorfindel yep, it wasn't relative at first. We caught it and fixed it quickly. You're not hallucinating a problem :) — Cesar M ♦ 16 secs ago
What artifical stupity will part 4 bring? Unfortunately, the days are long gone when SE cared what we thought — Starship is go for launch 22 secs ago
Unfortunately, I think the short answer for the "can they" is just "yes", because it's their site and they can decide to do effectively whatever they want with it, abiding by their own terms that they themselves define. I do think there's a legitimate argument for the mod agreement amendment bit, but I don't think anything outside of that is actually more than a violation of long standing convention rather than hard, legally-binding policy, in my understanding. — zcoop98 37 secs ago
This feels like it's roughly a duplicate of Please follow your own policies on making new policies, which covers a lot of the same ground. — zcoop98 44 secs ago
@starball I certainly have no intention of misquoting to mudsling. It was an honest mistake. I was quite freshly boiling at the moment I read through that blog post, so I misunderstood this part specifically. I went back, read through it again, and realized that you're completely right: I did misunderstand, and as a consequence, misrepresent that particular part. I apologize for that, and thank you very much for pointing it out. I appreciate it. I have removed that part from my post. — Andreas detests censorship 42 secs ago
@zcoop98 Of course they can, they also can make these sites into porn sharing or a forum or anything else. The question here is why they think they can impose something like that on us while still claiming SE sites are run by the communities. Chris's post is close, yes, but here I am asking for an official response on how the company feels it can square the mod agreement and the "run by us" bit with what they are doing. — terdon 1 min ago
@starball I'm not saying that everybody here gets nothing from the work we currently put into the site. Obviously, we do curation because of our affection for the site, and in return, we get a cleaner site that we find highly useful. Without curation, the site wouldn't serve its purpose, and the intended goals (of the community) could not have been fulfilled. My point is that it's not fair to use us as free labour to train their models, which they will then sell for other purposes. The ML does not benefit the community, and only a community benefit respects and acknowledges our free labour. — Andreas detests censorship 58 secs ago
pasting a comment I just made elsewhere: "When you made an account, you implicitly agreed to the ToS, which also gives SE the right to commercially exploit your content. Not that that brings me great joy, but it's a fact, and you need to accept it." — starball 16 secs ago
The fact is, living in a symbiotic relationship in which both sides get a mutual benefit, is the state we must aim for. What SE is doing now, and has been for a while, is turning themselves into a parasite, feeding off of us. — Andreas detests censorship 42 secs ago
@starball That comment misses the point of my criticism. I am fully aware that they can commercially exploit our contributions, and I never said otherwise. But I'm not willing to do it for free. If the community gets nothing in return, well, other than burdens and disrespectful treatment, there is no point for us to continue. Just because they wrote that in the ToS, doesn't mean they are morally in the clear. — Andreas detests censorship 29 secs ago
10:32 PM
As I explained to one user: if you take some chat bot's output an password every sentence (as this user claimed to be doing) the result is still not your own work; presenting it as your own work is still plagiarism. That user apologized in their mod message reply. We hardly ever get genuine apologies in mod message replies. — rob 13 secs ago
@rob I've also rarely contacted users about AI content and yet saw multiple apologies. — Bryan Krause 39 secs ago
"The company has never before presumed to dictate.": never before recent events, precisely. And I have no illusions about management caring, I am curious to see if they can provide justification. I don't expect I'll buy it if they do. — terdon 46 secs ago
Moving to leave open as this was tagged as under review by a staff member (not just a mod unlike the other one). — Sonic the Anonymous Hedgehog 8 secs ago
1 hour later…
11:49 PM
When is the "We've replaced our CEO" question coming. Thats what im waiting for. — Starship is go for launch 58 secs ago
At the moment that really is the company spirit @This_is_NOT_a_forum — Starship is go for launch 29 secs ago
00:00 - 15:0015:00 - 00:00
« first day (7 days earlier) ← previous day next day → last day (353 days later) »
Transcript for
May30
May '2331
Jun1
♦ Meta Stack Exchange Comment Archive
A comprehensive archive of all Meta Stack Exchange Comments. C...
